OpenAI launches Images 2.6 with GIFs, reference images, and Flare/Sunburst API models
On 4 Nov 2026 OpenAI released Images 2.6 in ChatGPT and Codex on all tiers, adding GIF creation and reusable reference images; the API adds GPT-Image-2.6 Flare and Sunburst, which OpenAI says are about 50% faster, with Flare as low as ~6 seconds at low quality.
Verified against OpenAI’s post Introducing Images 2.6 (dated November 4, 2026) and matching posts from @OpenAI / @OpenAIDevs:
OpenAI is launching Images 2.6 across ChatGPT, Codex, and the API, saying image generation is about 50% faster than before and giving more control across editing rounds.
For ChatGPT / Codex users
- Available globally today on all subscription tiers on desktop, web, and mobile.
- Create animated GIFs from a prompt or image; adjust speed and edit individual frames.
- Save reference images (e.g., people, pets, characters) for reuse without re-uploading.
For developers (API)
- New models: GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst (also in Codex).
- OpenAI says that with quality set to low, Flare can generate images in as little as six seconds and improves layout for posters and product imagery; Sunburst targets higher quality and stronger consistency across edits at medium or high quality.
- API additions include image markup (already in ChatGPT), SVG exports, and a complexity parameter that adjusts how much planning the model does; quality still controls visual detail. OpenAI recommends low complexity for most edits, and medium/high when scenes have many elements or constraints.
Pricing (official API table)
Per OpenAI API Pricing image-generation rates (per 1M tokens): gpt-image-2.6-sunburst and gpt-image-2.6-flare list Image input/output/cached input at $8 / $30 / $2, and Text input/cached input at $5 / $1.25. The post also points to pricing details and a system card for safety evaluations.
OpenAI also states that people create about 4 billion images per week across ChatGPT, Codex, and the API—on average more than half a billion per day. This item reports the product/API launch; customer quotes in the post are not treated as independently verified claims.